Chapter 23

Software Cost Estimation

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 1

Software cost estimation

 Predicting the resources required

for a software development
process

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 2

Objectives
 To introduce the fundamentals of software
costing and pricing
 To describe three metrics for software
productivity assessment
 To explain why different techniques should be
used for software estimation
 To describe the COCOMO 2 algorithmic cost
estimation model

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 3

Topics covered
 Productivity
 Estimation techniques
 Algorithmic cost modelling
 Project duration and staffing

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 4

Fundamental estimation questions
 How much effort is required to complete an
activity?
 How much calendar time is needed to complete
an activity?
 What is the total cost of an activity?
 Project estimation and scheduling and
interleaved management activities

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 5

Software cost components
 Hardware and software costs
 Travel and training costs
 Effort costs (the dominant factor in most
projects)
• salaries of engineers involved in the project
• Social and insurance costs
 Effort costs must take overheads into account
• costs of building, heating, lighting
• costs of networking and communications
• costs of shared facilities (e.g library, staff restaurant, etc.)

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 6

Costing and pricing
 Estimates are made to discover the cost, to the
developer, of producing a software system
 There is not a simple relationship between the
development cost and the price charged to the
customer
 Broader organisational, economic, political and
business considerations influence the price
charged

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 7

Software pricing factors

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 8

Programmer productivity
 A measure of the rate at which individual
engineers involved in software development
produce software and associated
documentation
 Not quality-oriented although quality assurance
is a factor in productivity assessment
 Essentially, we want to measure useful
functionality produced per time unit

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 9

Productivity measures
 Size related measures based on some output
from the software process. This may be lines of
delivered source code, object code instructions,
etc.
 Function-related measures based on an estimate
of the functionality of the delivered software.
Function-points are the best known of this type of
measure

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 10

Measurement problems
 Estimating the size of the measure
 Estimating the total number of programmer
months which have elapsed
 Estimating contractor productivity (e.g.
documentation team) and incorporating this
estimate in overall estimate

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 11

Lines of code
 What's a line of code?
• The measure was first proposed when programs were typed on
cards with one line per card
• How does this correspond to statements as in Java which can
span several lines or where there can be several statements on
one line
 What programs should be counted as part of the
system?
 Assumes linear relationship between system
size and volume of documentation

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 12

Productivity comparisons
 The lower level the language, the more
productive the programmer
• The same functionality takes more code to implement in a
lower-level language than in a high-level language
 The more verbose the programmer, the higher
the productivity
• Measures of productivity based on lines of code suggest that
programmers who write verbose code are more productive than
programmers who write compact code

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 13

High and low level languages

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 14

System development times

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 15

Function points
 Based on a combination of program
characteristics
• external inputs and outputs
• user interactions
• external interfaces
• files used by the system
 A weight is associated with each of these
 The function point count is computed by
multiplying each raw count by the weight and
summing all values

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 16

Function points
 Function point count modified by complexity of
the project
 FPs can be used to estimate LOC depending on
the average number of LOC per FP for a given
language
• LOC = AVC * number of function points
• AVC is a language-dependent factor varying from 200-300 for
assemble language to 2-40 for a 4GL
 FPs are very subjective. They depend on the
estimator.
• Automatic function-point counting is impossible
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 17
Object points
 Object points are an alternative function-related
measure to function points when 4Gls or similar
languages are used for development
 Object points are NOT the same as object
classes
 The number of object points in a program is a
weighted estimate of
• The number of separate screens that are displayed
• The number of reports that are produced by the system
• The number of 3GL modules that must be developed to
supplement the 4GL code
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 18
Object point estimation
 Object points are easier to estimate from a
specification than function points as they are
simply concerned with screens, reports and 3GL
modules
 They can therefore be estimated at an early point
in the development process. At this stage, it is
very difficult to estimate the number of lines of
code in a system

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 19

Productivity estimates
 Real-time embedded systems, 40-160
LOC/P-month
 Systems programs , 150-400 LOC/P-month
 Commercial applications, 200-800
LOC/P-month
 In object points, productivity has been measured
between 4 and 50 object points/month
depending on tool support and developer
capability

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 20

Factors affecting productivity

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 21

Quality and productivity
 All metrics based on volume/unit time are
flawed because they do not take quality into
account
 Productivity may generally be increased at the
cost of quality
 It is not clear how productivity/quality metrics
are related
 If change is constant then an approach based on
counting lines of code is not meaningful

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 22

Estimation techniques
 There is no simple way to make an accurate
estimate of the effort required to develop a
software system
• Initial estimates are based on inadequate information in a user
requirements definition
• The software may run on unfamiliar computers or use new
technology
• The people in the project may be unknown
 Project cost estimates may be self-fulfilling
• The estimate defines the budget and the product is adjusted to
meet the budget

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 23

Estimation techniques
 Algorithmic cost modelling
 Expert judgement
 Estimation by analogy
 Parkinson's Law
 Pricing to win

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 24

Algorithmic code modelling
 A formulaic approach based on historical cost
information and which is generally based on the
size of the software
 Discussed later in this chapter

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 25

Expert judgement
 One or more experts in both software
development and the application domain use
their experience to predict software costs.
Process iterates until some consensus is
reached.
 Advantages: Relatively cheap estimation
method. Can be accurate if experts have direct
experience of similar systems
 Disadvantages: Very inaccurate if there are no
experts!
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 26
Estimation by analogy
 The cost of a project is computed by comparing
the project to a similar project in the same
application domain
 Advantages: Accurate if project data available
 Disadvantages: Impossible if no comparable
project has been tackled. Needs systematically
maintained cost database

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 27

Parkinson's Law
 The project costs whatever resources are
available
 Advantages: No overspend
 Disadvantages: System is usually unfinished

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 28

Pricing to win
 The project costs whatever the customer has to
spend on it
 Advantages: You get the contract
 Disadvantages: The probability that the
customer gets the system he or she wants is
small. Costs do not accurately reflect the work
required

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 29

Top-down and bottom-up estimation
 Any of these approaches may be used top-down
or bottom-up
 Top-down
• Start at the system level and assess the overall system
functionality and how this is delivered through sub-systems
 Bottom-up
• Start at the component level and estimate the effort required for
each component. Add these efforts to reach a final estimate

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 30

Top-down estimation
 Usable without knowledge of the system
architecture and the components that might be
part of the system
 Takes into account costs such as integration,
configuration management and documentation
 Can underestimate the cost of solving difficult
low-level technical problems

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 31

Bottom-up estimation
 Usable when the architecture of the system is
known and components identified
 Accurate method if the system has been
designed in detail
 May underestimate costs of system level
activities such as integration and documentation

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 32

Estimation methods
 Each method has strengths and weaknesses
 Estimation should be based on several methods
 If these do not return approximately the same
result, there is insufficient information available
 Some action should be taken to find out more in
order to make more accurate estimates
 Pricing to win is sometimes the only applicable
method

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 33

Experience-based estimates
 Estimating is primarily experience-based
 However, new methods and technologies may
make estimating based on experience inaccurate
• Object oriented rather than function-oriented development
• Client-server systems rather than mainframe systems
• Off the shelf components
• Component-based software engineering
• CASE tools and program generators

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 34

Pricing to win
 This approach may seem unethical and
unbusinesslike
 However, when detailed information is lacking it
may be the only appropriate strategy
 The project cost is agreed on the basis of an
outline proposal and the development is
constrained by that cost
 A detailed specification may be negotiated or an
evolutionary approach used for system
development
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 35
Algorithmic cost modelling
 Cost is estimated as a mathematical function of
product, project and process attributes whose
values are estimated by project managers
• Effort = A  SizeB  M
• A is an organisation-dependent constant, B reflects the
disproportionate effort for large projects and M is a multiplier
reflecting product, process and people attributes

 Most commonly used product attribute for cost

estimation is code size
 Most models are basically similar but with
different values for A, B and M
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 36
Estimation accuracy
 The size of a software system can only be known
accurately when it is finished
 Several factors influence the final size
• Use of COTS and components
• Programming language
• Distribution of system
 As the development process progresses then the
size estimate becomes more accurate

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 37

Estimate uncertainty

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 38

The COCOMO model
 An empirical model based on project experience
 Well-documented, ‘independent’ model which is
not tied to a specific software vendor
 Long history from initial version published in
1981 (COCOMO-81) through various
instantiations to COCOMO 2
 COCOMO 2 takes into account different
approaches to software development, reuse, etc.

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 39

COCOMO 81

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 40

COCOMO 2 levels
 COCOMO 2 is a 3 level model that allows
increasingly detailed estimates to be prepared
as development progresses
 Early prototyping level
• Estimates based on object points and a simple formula is used
for effort estimation
 Early design level
• Estimates based on function points that are then translated to
LOC
 Post-architecture level
• Estimates based on lines of source code

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 41

Early prototyping level
 Supports prototyping projects and projects where
there is extensive reuse
 Based on standard estimates of developer
productivity in object points/month
 Takes CASE tool use into account
 Formula is
• PM = ( NOP  (1 - %reuse/100 ) ) / PROD
• PM is the effort in person-months, NOP is the number of object
points and PROD is the productivity

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 42

Object point productivity

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 43

Early design level
 Estimates can be made after the requirements
have been agreed
 Based on standard formula for algorithmic
models
• PM = A  SizeB  M + PMm where
• M = PERS  RCPX  RUSE  PDIF  PREX  FCIL  SCED
• PMm = (ASLOC  (AT/100)) / ATPROD
• A = 2.5 in initial calibration, Size in KLOC, B varies from 1.1 to
1.24 depending on novelty of the project, development flexibility,
risk management approaches and the process maturity

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 44

Multipliers
 Multipliers reflect the capability of the developers,
the non-functional requirements, the familiarity
with the development platform, etc.
• RCPX - product reliability and complexity
• RUSE - the reuse required
• PDIF - platform difficulty
• PREX - personnel experience
• PERS - personnel capability
• SCED - required schedule
• FCIL - the team support facilities
 PM reflects the amount of automatically generated
code
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 45
Post-architecture level
 Uses same formula as early design estimates
 Estimate of size is adjusted to take into account
• Requirements volatility. Rework required to support change
• Extent of possible reuse. Reuse is non-linear and has
associated costs so this is not a simple reduction in LOC
• ESLOC = ASLOC  (AA + SU +0.4DM + 0.3CM +0.3IM)/100
» ESLOC is equivalent number of lines of new code. ASLOC is the
number of lines of reusable code which must be modified, DM is
the percentage of design modified, CM is the percentage of the
code that is modified , IM is the percentage of the original
integration effort required for integrating the reused software.
» SU is a factor based on the cost of software understanding, AA is
a factor which reflects the initial assessment costs of deciding if
software may be reused.
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 46
The exponent term
 This depends on 5 scale factors (see next slide).
Their sum/100 is added to 1.01
 Example
• Precedenteness - new project - 4
• Development flexibility - no client involvement - Very high - 1
• Architecture/risk resolution - No risk analysis - V. Low - 5
• Team cohesion - new team - nominal - 3
• Process maturity - some control - nominal - 3
 Scale factor is therefore 1.17

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 47

Exponent scale factors

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 48

Multipliers
 Product attributes
• concerned with required characteristics of the software product being
developed
 Computer attributes
• constraints imposed on the software by the hardware platform

 Personnel attributes
• multipliers that take the experience and capabilities of the people
working on the project into account.

 Project attributes
• concerned with the particular characteristics of the software
development project

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 49

Project cost drivers

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 50

Effects of cost drivers

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 51

Project planning
 Algorithmic cost models provide a basis for
project planning as they allow alternative
strategies to be compared
 Embedded spacecraft system
• Must be reliable
• Must minimise weight (number of chips)
• Multipliers on reliability and computer constraints > 1
 Cost components
• Target hardware
• Development platform
• Effort required

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 52

Management options

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 53

Management options costs

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 54

Option choice
 Option D (use more experienced staff) appears
to be the best alternative
• However, it has a high associated risk as expreienced staff may
be difficult to find
 Option C (upgrade memory) has a lower cost
saving but very low risk
 Overall, the model reveals the importance of staff
experience in software development

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 55

Project duration and staffing
 As well as effort estimation, managers must
estimate the calendar time required to complete
a project and when staff will be required
 Calendar time can be estimated using a
COCOMO 2 formula
• TDEV = 3  (PM)(0.33+0.2*(B-1.01))
• PM is the effort computation and B is the exponent computed
as discussed above (B is 1 for the early prototyping model).
This computation predicts the nominal schedule for the project
 The time required is independent of the number
of people working on the project
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 56
Staffing requirements
 Staff required can’t be computed by diving the
development time by the required schedule
 The number of people working on a project
varies depending on the phase of the project
 The more people who work on the project, the
more total effort is usually required
 A very rapid build-up of people often correlates
with schedule slippage

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 57

Key points
 Factors affecting productivity include individual
aptitude, domain experience, the development
project, the project size, tool support and the
working environment
 Different techniques of cost estimation should be
used when estimating costs
 Software may be priced to gain a contract and
the functionality adjusted to the price

©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 58

Key points
 Algorithmic cost estimation is difficult because
of the need to estimate attributes of the finished
product
 The COCOMO model takes project, product,
personnel and hardware attributes into account
when predicting effort required
 Algorithmic cost models support quantitative
option analysis
 The time to complete a project is not
proportional to the number of people working
on the project
©Ian Sommerville 2000 Software Engineering, 6th edition. Chapter 23 Slide 59